FAMILY MEMBERS OF OLDER PATIENTS CONSIDERING TREATMENT OPTIONS FOR KIDNEY FAILURE DO NOT MAKE DIFFERENT TRADE-OFFS BETWEEN TREATMENT BENEFITS AND BURDENS THAN THE PATIENTS THEMSELVES

Introduction:

Older, frailer people with multiple conditions at risk of kidney failure typically choose between treatment with dialysis and conservative kidney management (CKM). Family members influence and are affected by treatment decisions and have been reported to favour more ‘active’ and life-sustaining treatment options than patients. However, quantitative comparisons between patient and family preferences for treatments are scarce and have not focussed on the older population, for whom the survival and quality of life benefits of dialysis are uncertain. We report analysis from a choice experiment measuring the preferences of older patients and their family members for the characteristics of dialysis and CKM. We hypothesised that family members would require less survival benefit to accept the burdens associated with dialysis and would be less willing to trade-off survival to preserve independence.

Methods:

A choice experiment informed by qualitative research was conducted with UK over-65-year-olds with eGFR≤20mls/min/1.73m2. Patients were asked to invite one family member involved in treatment decision-making, who completed the same questions. A mixed logit analysis quantified trade-offs between frequency and location of treatments, survival, and capability (the ability to do important activities). No adjustment was made for clinical or sociodemographic factors. Scenarios and attributes are depicted in Figure and Table 1. Patients and family members were treated as dyads in the analysis, with an interaction term used to examine whether patient/family member status influenced preferences.

Results:

314 participants constituting 157 patient-family member dyads took part (median patient age 74years, eGFR 13.5mls/min/1.73m2). Reduced treatment frequency, improved capability, and better survival were strongly associated with choice (p<0.001). Home vs. hospital treatment did not reach the threshold for statistical significance (p=0.086). Overall, participants needed a 5% survival benefit two years after starting treatment to accept one additional treatment per week, and a 14% survival benefit two years after starting treatment to prevent a halving of their capability. No differences in the trade-offs made by patients and their family members reached the threshold for statistical significance (Table 2).

Conclusions:

Stated preferences indicated participants favoured higher survival probabilities, but only if their capability was preserved and the frequency of care was acceptable. Location of care did not appear to be valued. We found no evidence to support our hypothesis that family members would make quantitatively different trade-offs from patients. Findings thus far do not support the concept that family members favour more active treatments because of stronger relative preferences for survival. Analyses accounting for clinical and sociodemographic factors are underway to examine whether subgroups of patients, family members, or dyads exhibit different preferences – this may unveil patterns indecipherable in these ‘average’ findings. We will also be able to compare the preferences of this patient group with approximately 200 patient participants who did not recruit a family member. Further mixed-methods research will explore if and how preferences change with time and disease progression, and how stated preferences relate to treatment choice.

I have no potential conflict of interest to disclose.

I did not use generative AI and AI-assisted technologies in the writing process.